/**
* Returns the possible neutral losses expected by default for a given peptide. /!\ this method
* will work only if the PTM found in the peptide are in the PTMFactory.
*
* @param spectrumIdentificationAssumption the spectrumIdentificationAssumption of interest
* @param sequenceMatchingPreferences the sequence matching settings for peptide to protein
* mapping
* @param ptmSequenceMatchingPreferences the sequence matching settings for PTM to peptide mapping
* @return the expected possible neutral losses
* @throws IOException exception thrown whenever an error occurred while interacting with a file
* while mapping potential modification sites
* @throws InterruptedException exception thrown whenever a threading issue occurred while mapping
* potential modification sites
* @throws ClassNotFoundException exception thrown whenever an error occurred while deserializing
* an object from the ProteinTree
* @throws SQLException exception thrown whenever an error occurred while interacting with the
* ProteinTree
*/
public static NeutralLossesMap getDefaultLosses(
SpectrumIdentificationAssumption spectrumIdentificationAssumption,
SequenceMatchingPreferences sequenceMatchingPreferences,
SequenceMatchingPreferences ptmSequenceMatchingPreferences)
throws IOException, InterruptedException, ClassNotFoundException, SQLException {
if (spectrumIdentificationAssumption instanceof PeptideAssumption) {
PeptideAssumption peptideAssumption = (PeptideAssumption) spectrumIdentificationAssumption;
return PeptideSpectrumAnnotator.getDefaultLosses(
peptideAssumption.getPeptide(),
sequenceMatchingPreferences,
ptmSequenceMatchingPreferences);
} else if (spectrumIdentificationAssumption instanceof TagAssumption) {
TagAssumption tagAssumption = (TagAssumption) spectrumIdentificationAssumption;
return TagSpectrumAnnotator.getDefaultLosses(
tagAssumption.getTag(), ptmSequenceMatchingPreferences);
} else {
throw new IllegalArgumentException(
"Default neutral loss map not implemented for SpectrumIdentificationAssumption "
+ spectrumIdentificationAssumption.getClass()
+ ".");
}
}
/**
* Fills the PSM specific map.
*
* @param inputMap The input map
* @param waitingHandler the handler displaying feedback to the user
* @param shotgunProtocol information about the protocol
* @param identificationParameters the identification parameters
* @throws java.sql.SQLException exception thrown whenever an error occurred while interacting
* with the back-end database
* @throws java.io.IOException exception thrown whenever an error occurred while reading an
* external file
* @throws java.lang.ClassNotFoundException exception thrown whenever an error occurred while
* deserializing an object
* @throws java.lang.InterruptedException exception thrown whenever a threading error occurred
* @throws uk.ac.ebi.jmzml.xml.io.MzMLUnmarshallerException exception thrown whenever an error
* occurred while reading an mzML file
*/
public void selectBestHitAndFillPsmMap(
InputMap inputMap,
WaitingHandler waitingHandler,
ShotgunProtocol shotgunProtocol,
IdentificationParameters identificationParameters)
throws SQLException, IOException, ClassNotFoundException, InterruptedException,
MzMLUnmarshallerException {
waitingHandler.setSecondaryProgressCounterIndeterminate(false);
waitingHandler.setMaxSecondaryProgressCounter(identification.getSpectrumIdentificationSize());
PeptideSpectrumAnnotator spectrumAnnotator = new PeptideSpectrumAnnotator();
boolean multiSE = inputMap.isMultipleAlgorithms();
SequenceMatchingPreferences sequenceMatchingPreferences =
identificationParameters.getSequenceMatchingPreferences();
AnnotationSettings annotationPreferences = identificationParameters.getAnnotationPreferences();
PeptideAssumptionFilter idFilter = identificationParameters.getPeptideAssumptionFilter();
// Keep a map of the spectrum keys grouped by peptide
HashMap<String, ArrayList<String>> orderedPsmMap = null;
if (MemoryConsumptionStatus.memoryUsed() < 0.8) {
orderedPsmMap =
new HashMap<String, ArrayList<String>>(
identification.getSpectrumIdentificationMap().size());
}
PSParameter psParameter = new PSParameter();
for (String spectrumFileName : identification.getSpectrumFiles()) {
HashMap<String, ArrayList<String>> keysMap = null;
if (orderedPsmMap != null) {
keysMap = new HashMap<String, ArrayList<String>>();
}
PsmIterator psmIterator =
identification.getPsmIterator(spectrumFileName, null, true, waitingHandler);
while (psmIterator.hasNext()) {
SpectrumMatch advocateMatch = psmIterator.next();
String spectrumKey = advocateMatch.getKey();
// map of the peptide first hits for this spectrum: score -> max protein count -> max search
// engine votes -> amino acids annotated -> min mass deviation -> peptide sequence
HashMap<
Double,
HashMap<
Integer,
HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>>>
peptideAssumptions =
new HashMap<
Double,
HashMap<
Integer,
HashMap<
Integer,
HashMap<
Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>>>();
// map of the tag first hits: score -> assumptions
HashMap<Double, ArrayList<TagAssumption>> tagAssumptions =
new HashMap<Double, ArrayList<TagAssumption>>();
ArrayList<String> identifications = new ArrayList<String>();
HashMap<Integer, HashMap<Double, ArrayList<SpectrumIdentificationAssumption>>> assumptions =
identification.getAssumptions(spectrumKey);
for (int searchEngine1 : assumptions.keySet()) {
HashMap<Double, ArrayList<SpectrumIdentificationAssumption>> advocate1Map =
assumptions.get(searchEngine1);
double bestEvalue = Collections.min(advocate1Map.keySet());
for (SpectrumIdentificationAssumption assumption1 : advocate1Map.get(bestEvalue)) {
if (assumption1 instanceof PeptideAssumption) {
PeptideAssumption peptideAssumption1 = (PeptideAssumption) assumption1;
String id = peptideAssumption1.getPeptide().getKey();
if (!identifications.contains(id)) {
psParameter = (PSParameter) peptideAssumption1.getUrParam(psParameter);
double p;
if (multiSE && sequenceFactory.concatenatedTargetDecoy()) {
p = psParameter.getSearchEngineProbability();
} else {
p = peptideAssumption1.getScore();
}
int nSE = 1;
int proteinMax = 1;
for (String protein :
peptideAssumption1
.getPeptide()
.getParentProteins(sequenceMatchingPreferences)) {
Integer tempCount = proteinCount.get(protein);
if (tempCount != null && tempCount > proteinMax) {
proteinMax = tempCount;
}
}
for (int searchEngine2 : assumptions.keySet()) {
if (searchEngine1 != searchEngine2) {
HashMap<Double, ArrayList<SpectrumIdentificationAssumption>> advocate2Map =
assumptions.get(searchEngine2);
boolean found = false;
ArrayList<Double> eValues2 = new ArrayList<Double>(advocate2Map.keySet());
Collections.sort(eValues2);
for (double eValue2 : eValues2) {
for (SpectrumIdentificationAssumption assumption2 :
advocate2Map.get(eValue2)) {
if (assumption2 instanceof PeptideAssumption) {
PeptideAssumption peptideAssumption2 = (PeptideAssumption) assumption2;
if (peptideAssumption1
.getPeptide()
.isSameSequenceAndModificationStatus(
peptideAssumption2.getPeptide(), sequenceMatchingPreferences)) {
PSParameter psParameter2 =
(PSParameter) peptideAssumption2.getUrParam(psParameter);
p = p * psParameter2.getSearchEngineProbability();
nSE++;
found = true;
break;
}
}
}
if (found) {
break;
}
}
}
}
identifications.add(id);
HashMap<
Integer,
HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>>
pMap = peptideAssumptions.get(p);
if (pMap == null) {
pMap =
new HashMap<
Integer,
HashMap<
Integer,
HashMap<
Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>>(
advocate1Map.size());
peptideAssumptions.put(p, pMap);
}
HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>
proteinMaxMap = pMap.get(proteinMax);
if (proteinMaxMap == null) {
proteinMaxMap =
new HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>(1);
pMap.put(proteinMax, proteinMaxMap);
}
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>> nSeMap =
proteinMaxMap.get(nSE);
if (nSeMap == null) {
nSeMap =
new HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>(1);
proteinMaxMap.put(nSE, nSeMap);
HashMap<Double, HashMap<String, PeptideAssumption>> coverageMap =
new HashMap<Double, HashMap<String, PeptideAssumption>>(1);
nSeMap.put(-1, coverageMap);
HashMap<String, PeptideAssumption> assumptionMap =
new HashMap<String, PeptideAssumption>(1);
coverageMap.put(-1.0, assumptionMap);
assumptionMap.put(
peptideAssumption1.getPeptide().getSequenceWithLowerCasePtms(),
peptideAssumption1);
} else {
MSnSpectrum spectrum = (MSnSpectrum) spectrumFactory.getSpectrum(spectrumKey);
HashMap<Double, HashMap<String, PeptideAssumption>> coverageMap = nSeMap.get(-1);
if (coverageMap != null) {
HashMap<String, PeptideAssumption> assumptionMap = coverageMap.get(-1.0);
for (PeptideAssumption tempAssumption :
assumptionMap.values()) { // There should be only one
Peptide peptide = tempAssumption.getPeptide();
SpecificAnnotationSettings specificAnnotationPreferences =
annotationPreferences.getSpecificAnnotationPreferences(
spectrum.getSpectrumKey(),
tempAssumption,
identificationParameters.getSequenceMatchingPreferences(),
identificationParameters
.getPtmScoringPreferences()
.getSequenceMatchingPreferences());
HashMap<Integer, ArrayList<IonMatch>> coveredAminoAcids =
spectrumAnnotator.getCoveredAminoAcids(
annotationPreferences,
specificAnnotationPreferences,
(MSnSpectrum) spectrum,
peptide);
int nIons = coveredAminoAcids.size();
nSeMap.put(nIons, coverageMap);
}
nSeMap.remove(-1);
}
Peptide peptide = peptideAssumption1.getPeptide();
SpecificAnnotationSettings specificAnnotationPreferences =
annotationPreferences.getSpecificAnnotationPreferences(
spectrum.getSpectrumKey(),
peptideAssumption1,
identificationParameters.getSequenceMatchingPreferences(),
identificationParameters
.getPtmScoringPreferences()
.getSequenceMatchingPreferences());
HashMap<Integer, ArrayList<IonMatch>> coveredAminoAcids =
spectrumAnnotator.getCoveredAminoAcids(
annotationPreferences,
specificAnnotationPreferences,
(MSnSpectrum) spectrum,
peptide);
int nIons = coveredAminoAcids.size();
coverageMap = nSeMap.get(nIons);
if (coverageMap == null) {
coverageMap = new HashMap<Double, HashMap<String, PeptideAssumption>>(1);
HashMap<String, PeptideAssumption> assumptionMap =
new HashMap<String, PeptideAssumption>(1);
assumptionMap.put(
peptideAssumption1.getPeptide().getSequenceWithLowerCasePtms(),
peptideAssumption1);
coverageMap.put(-1.0, assumptionMap);
nSeMap.put(nIons, coverageMap);
} else {
HashMap<String, PeptideAssumption> assumptionMap = coverageMap.get(-1.0);
if (assumptionMap != null) {
for (PeptideAssumption tempAssumption :
assumptionMap.values()) { // There should be only one
double massError =
Math.abs(
tempAssumption.getDeltaMass(
spectrum.getPrecursor().getMz(),
shotgunProtocol.isMs1ResolutionPpm()));
coverageMap.put(massError, assumptionMap);
}
coverageMap.remove(-1.0);
}
double massError =
Math.abs(
peptideAssumption1.getDeltaMass(
spectrum.getPrecursor().getMz(),
shotgunProtocol.isMs1ResolutionPpm()));
assumptionMap = coverageMap.get(massError);
if (assumptionMap == null) {
assumptionMap = new HashMap<String, PeptideAssumption>(1);
coverageMap.put(massError, assumptionMap);
}
assumptionMap.put(
peptideAssumption1.getPeptide().getSequenceWithLowerCasePtms(),
peptideAssumption1);
}
}
}
} else if (assumption1 instanceof TagAssumption) {
TagAssumption tagAssumption = (TagAssumption) assumption1;
ArrayList<TagAssumption> assumptionList = tagAssumptions.get(bestEvalue);
if (assumptionList == null) {
assumptionList = new ArrayList<TagAssumption>();
tagAssumptions.put(bestEvalue, assumptionList);
}
assumptionList.add(tagAssumption);
}
}
}
SpectrumMatch spectrumMatch = new SpectrumMatch(spectrumKey);
if (!peptideAssumptions.isEmpty()) {
PeptideAssumption bestPeptideAssumption = null;
ArrayList<Double> ps = new ArrayList<Double>(peptideAssumptions.keySet());
Collections.sort(ps);
double retainedP = 0;
for (double p : ps) {
retainedP = p;
HashMap<
Integer,
HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>>
pMap = peptideAssumptions.get(p);
ArrayList<Integer> proteinMaxs = new ArrayList<Integer>(pMap.keySet());
Collections.sort(proteinMaxs, Collections.reverseOrder());
for (int proteinMax : proteinMaxs) {
HashMap<
Integer,
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>>>
proteinMaxMap = pMap.get(proteinMax);
ArrayList<Integer> nSEs = new ArrayList<Integer>(proteinMaxMap.keySet());
Collections.sort(nSEs, Collections.reverseOrder());
for (int nSE : nSEs) {
HashMap<Integer, HashMap<Double, HashMap<String, PeptideAssumption>>> nSeMap =
proteinMaxMap.get(nSE);
ArrayList<Integer> coverages = new ArrayList<Integer>(nSeMap.keySet());
Collections.sort(coverages, Collections.reverseOrder());
for (Integer coverage : coverages) {
HashMap<Double, HashMap<String, PeptideAssumption>> coverageMap =
nSeMap.get(coverage);
ArrayList<Double> minErrors = new ArrayList<Double>(coverageMap.keySet());
Collections.sort(minErrors);
for (double minError : minErrors) {
HashMap<String, PeptideAssumption> bestPeptideAssumptions =
coverageMap.get(minError);
ArrayList<String> sequences =
new ArrayList<String>(bestPeptideAssumptions.keySet());
Collections.sort(sequences);
for (String sequence : sequences) {
PeptideAssumption peptideAssumption = bestPeptideAssumptions.get(sequence);
if (idFilter.validateProteins(
peptideAssumption.getPeptide(), sequenceMatchingPreferences)) {
bestPeptideAssumption = peptideAssumption;
break;
}
}
if (bestPeptideAssumption != null) {
break;
}
}
if (bestPeptideAssumption != null) {
break;
}
}
if (bestPeptideAssumption != null) {
break;
}
}
if (bestPeptideAssumption != null) {
break;
}
}
if (bestPeptideAssumption != null) {
break;
}
}
if (bestPeptideAssumption != null) {
if (multiSE) {
// try to find the most likely modification localization based on the search engine
// results
HashMap<PeptideAssumption, ArrayList<Double>> assumptionPEPs =
new HashMap<PeptideAssumption, ArrayList<Double>>();
String bestAssumptionKey =
bestPeptideAssumption.getPeptide().getMatchingKey(sequenceMatchingPreferences);
for (int searchEngine : assumptions.keySet()) {
boolean found = false;
HashMap<Double, ArrayList<SpectrumIdentificationAssumption>> advocateMap =
assumptions.get(searchEngine);
ArrayList<Double> eValues = new ArrayList<Double>(advocateMap.keySet());
Collections.sort(eValues);
for (double eValue : eValues) {
for (SpectrumIdentificationAssumption assumption : advocateMap.get(eValue)) {
if (assumption instanceof PeptideAssumption) {
PeptideAssumption peptideAssumption = (PeptideAssumption) assumption;
if (peptideAssumption
.getPeptide()
.getMatchingKey(sequenceMatchingPreferences)
.equals(bestAssumptionKey)) {
found = true;
boolean found2 = false;
for (PeptideAssumption assumption1 : assumptionPEPs.keySet()) {
if (assumption1
.getPeptide()
.sameModificationsAs(peptideAssumption.getPeptide())) {
found2 = true;
psParameter = (PSParameter) assumption.getUrParam(psParameter);
ArrayList<Double> peps = assumptionPEPs.get(assumption1);
peps.add(psParameter.getSearchEngineProbability());
break;
}
}
if (!found2) {
ArrayList<Double> peps = new ArrayList<Double>(1);
assumptionPEPs.put(peptideAssumption, peps);
psParameter = (PSParameter) assumption.getUrParam(psParameter);
peps.add(psParameter.getSearchEngineProbability());
}
}
}
}
if (found) {
break;
}
}
}
Double bestSeP = null;
int nSe = -1;
for (PeptideAssumption peptideAssumption : assumptionPEPs.keySet()) {
ArrayList<Double> peps = assumptionPEPs.get(peptideAssumption);
Double sep = Collections.min(peps);
if (bestSeP == null || bestSeP > sep) {
bestSeP = sep;
nSe = peps.size();
bestPeptideAssumption = peptideAssumption;
} else if (peps.size() > nSe) {
if (sep != null && (Math.abs(sep - bestSeP) <= 1e-10)) {
nSe = peps.size();
bestPeptideAssumption = peptideAssumption;
}
}
}
}
// create a PeptideShaker match based on the best search engine match
Peptide sePeptide = bestPeptideAssumption.getPeptide();
ArrayList<String> psProteins =
new ArrayList<String>(sePeptide.getParentProteins(sequenceMatchingPreferences));
ArrayList<ModificationMatch> psModificationMatches = null;
if (sePeptide.isModified()) {
psModificationMatches =
new ArrayList<ModificationMatch>(sePeptide.getNModifications());
for (ModificationMatch seModMatch : sePeptide.getModificationMatches()) {
psModificationMatches.add(
new ModificationMatch(
seModMatch.getTheoreticPtm(),
seModMatch.isVariable(),
seModMatch.getModificationSite()));
}
}
Peptide psPeptide = new Peptide(sePeptide.getSequence(), psModificationMatches);
psPeptide.setParentProteins(psProteins);
PeptideAssumption psAssumption =
new PeptideAssumption(
psPeptide,
1,
Advocate.peptideShaker.getIndex(),
bestPeptideAssumption.getIdentificationCharge(),
retainedP);
spectrumMatch.setBestPeptideAssumption(psAssumption);
if (orderedPsmMap != null) {
String peptideKey = psPeptide.getMatchingKey(sequenceMatchingPreferences);
ArrayList<String> spectrumKeys = keysMap.get(peptideKey);
if (spectrumKeys == null) {
spectrumKeys = new ArrayList<String>();
keysMap.put(peptideKey, spectrumKeys);
}
spectrumKeys.add(spectrumKey);
}
psParameter = new PSParameter();
psParameter.setSpectrumProbabilityScore(retainedP);
PSParameter matchParameter =
(PSParameter) bestPeptideAssumption.getUrParam(psParameter);
psParameter.setSearchEngineProbability(matchParameter.getSearchEngineProbability());
psParameter.setAlgorithmDeltaPEP(matchParameter.getAlgorithmDeltaPEP());
psParameter.setDeltaPEP(matchParameter.getDeltaPEP());
matchesValidator
.getPsmMap()
.addPoint(
psParameter.getPsmProbabilityScore(),
spectrumMatch,
sequenceMatchingPreferences);
psParameter.setSpecificMapKey(
spectrumMatch.getBestPeptideAssumption().getIdentificationCharge().value + "");
identification.addSpectrumMatchParameter(spectrumKey, psParameter);
identification.updateSpectrumMatch(spectrumMatch);
}
}
if (!tagAssumptions.isEmpty()) {
ArrayList<Double> evalues = new ArrayList<Double>(tagAssumptions.keySet());
Double bestEvalue = Collections.min(evalues);
TagAssumption bestAssumption = tagAssumptions.get(bestEvalue).get(0);
spectrumMatch.setBestTagAssumption(bestAssumption);
identification.updateSpectrumMatch(spectrumMatch);
if (spectrumMatch.getBestPeptideAssumption() == null) {
psParameter = new PSParameter();
if (!multiSE) {
psParameter.setSpectrumProbabilityScore(bestEvalue);
}
PSParameter matchParameter = (PSParameter) bestAssumption.getUrParam(psParameter);
psParameter.setSearchEngineProbability(matchParameter.getSearchEngineProbability());
psParameter.setAlgorithmDeltaPEP(matchParameter.getAlgorithmDeltaPEP());
psParameter.setDeltaPEP(matchParameter.getDeltaPEP());
psParameter.setSpecificMapKey(
spectrumMatch.getBestTagAssumption().getIdentificationCharge().value + "");
identification.addSpectrumMatchParameter(spectrumKey, psParameter);
}
}
waitingHandler.increaseSecondaryProgressCounter();
if (waitingHandler.isRunCanceled()) {
return;
}
}
if (orderedPsmMap != null) {
ArrayList<String> orderedKeys =
new ArrayList<String>(
identification.getSpectrumIdentification(spectrumFileName).size());
for (ArrayList<String> keys : keysMap.values()) {
orderedKeys.addAll(keys);
}
orderedPsmMap.put(spectrumFileName, orderedKeys);
if (MemoryConsumptionStatus.memoryUsed() > 0.9) {
orderedPsmMap = null;
}
}
}
if (orderedPsmMap != null) {
metrics.setOrderedSpectrumKeys(orderedPsmMap);
}
// the protein count map is no longer needed
proteinCount.clear();
waitingHandler.setSecondaryProgressCounterIndeterminate(true);
}
public void testDB()
throws SQLException, IOException, ClassNotFoundException, SQLException,
ClassNotFoundException, InterruptedException {
String path = this.getClass().getResource("IdentificationDBTest.class").getPath();
path = path.substring(1, path.indexOf("/target/"));
path += "/src/test/resources/experiment/testDB";
try {
ObjectsCache cache = new ObjectsCache();
cache.setAutomatedMemoryManagement(false);
cache.setCacheSize(0);
IdentificationDB idDB = new IdentificationDB(path, "testId", true, cache);
try {
String spectrumKey = "spectrum_file_cus_spectrum_title";
String peptideKey = "PEPTIDE";
String proteinKey = "test_protein";
SpectrumMatch testSpectrumMatch = new SpectrumMatch(spectrumKey);
ArrayList<String> testProteins = new ArrayList<String>();
testProteins.add("test protein1");
testProteins.add("test protein2");
Peptide peptide = new Peptide(peptideKey, new ArrayList<ModificationMatch>());
peptide.setParentProteins(testProteins);
testSpectrumMatch.addHit(
Advocate.mascot.getIndex(),
new PeptideAssumption(
peptide, 1, Advocate.mascot.getIndex(), new Charge(Charge.PLUS, 2), 0.1, "no file"),
false);
idDB.addSpectrumMatch(testSpectrumMatch);
peptide = new Peptide(peptideKey, new ArrayList<ModificationMatch>());
peptide.setParentProteins(testProteins);
PeptideMatch testPeptideMatch = new PeptideMatch(peptide, peptide.getKey());
idDB.addPeptideMatch(testPeptideMatch);
ProteinMatch testProteinMatch = new ProteinMatch(proteinKey);
idDB.addProteinMatch(testProteinMatch);
testSpectrumMatch = idDB.getSpectrumMatch(spectrumKey, true);
Assert.assertTrue(testSpectrumMatch.getKey().equals(spectrumKey));
HashMap<Integer, HashMap<Double, ArrayList<SpectrumIdentificationAssumption>>>
assumptionsMap = testSpectrumMatch.getAssumptionsMap();
HashMap<Double, ArrayList<SpectrumIdentificationAssumption>> mascotAssumptions =
assumptionsMap.get(Advocate.mascot.getIndex());
Assert.assertTrue(mascotAssumptions.size() == 1);
ArrayList<Double> mascotScores = new ArrayList<Double>(mascotAssumptions.keySet());
Assert.assertTrue(mascotScores.size() == 1);
double bestScore = mascotScores.get(0);
Assert.assertTrue(bestScore == 0.1);
ArrayList<SpectrumIdentificationAssumption> bestAssumptions =
mascotAssumptions.get(bestScore);
PeptideAssumption bestAssumption = (PeptideAssumption) bestAssumptions.get(0);
Peptide bestPeptide = bestAssumption.getPeptide();
Assert.assertTrue(bestPeptide.getParentProteinsNoRemapping().size() == 2);
Assert.assertTrue(
bestPeptide.getParentProteinsNoRemapping().get(0).equals(testProteins.get(0)));
Assert.assertTrue(
bestPeptide.getParentProteinsNoRemapping().get(1).equals(testProteins.get(1)));
ArrayList<String> proteins = new ArrayList<String>();
proteins.add(proteinKey);
bestPeptide.setParentProteins(proteins);
idDB.updateMatch(testSpectrumMatch);
testSpectrumMatch = idDB.getSpectrumMatch(spectrumKey, true);
assumptionsMap = testSpectrumMatch.getAssumptionsMap();
mascotAssumptions = assumptionsMap.get(Advocate.mascot.getIndex());
Assert.assertTrue(mascotAssumptions.size() == 1);
mascotScores = new ArrayList<Double>(mascotAssumptions.keySet());
Assert.assertTrue(mascotScores.size() == 1);
bestScore = mascotScores.get(0);
Assert.assertTrue(bestScore == 0.1);
bestAssumptions = mascotAssumptions.get(bestScore);
bestAssumption = (PeptideAssumption) bestAssumptions.get(0);
bestPeptide = bestAssumption.getPeptide();
Assert.assertTrue(bestPeptide.getParentProteinsNoRemapping().size() == 1);
Assert.assertTrue(bestPeptide.getParentProteinsNoRemapping().get(0).equals(proteinKey));
testPeptideMatch = idDB.getPeptideMatch(peptideKey, true);
Assert.assertTrue(testPeptideMatch.getKey().equals(peptideKey));
testProteinMatch = idDB.getProteinMatch(proteinKey, true);
Assert.assertTrue(testProteinMatch.getKey().equals(proteinKey));
double testScore = 12.3;
PepnovoAssumptionDetails testParameter = new PepnovoAssumptionDetails();
testParameter.setRankScore(testScore);
idDB.addSpectrumMatchParameter(spectrumKey, testParameter);
testParameter =
(PepnovoAssumptionDetails)
idDB.getSpectrumMatchParameter(spectrumKey, testParameter, true);
Assert.assertTrue(testParameter.getRankScore() == testScore);
} finally {
idDB.close();
}
} finally {
File dbFolder = new File(path);
DerbyUtil.closeConnection();
Util.deleteDir(dbFolder);
}
}
/**
* Selects a first hit in a list of equally scoring peptide matches. The selection is made based
* on: 1 - The occurrence of the protein detection as given in the proteinCount map 2 - The
* sequence coverage by fragment ions 3 - The precursor mass error.
*
* <p>If no best hit is found, the first one sorted alphabetically is retained.
*
* @param spectrumKey the key of the spectrum
* @param firstHits list of equally scoring peptide matches
* @param proteinCount map of the number of peptides for every protein
* @param sequenceMatchingPreferences the sequence matching preferences
* @param shotgunProtocol the shotgun protocol
* @param identificationParameters the identification parameters
* @param spectrumAnnotator the spectrum annotator to use
* @return a first hit from the list of equally scoring peptide matches
* @throws IOException exception thrown whenever an IO exception occurred while remapping the
* proteins or getting the spectrum
* @throws InterruptedException exception thrown whenever an interrupted exception occurred while
* remapping the proteins or getting the spectrum
* @throws SQLException exception thrown whenever an SQL exception occurred while interacting with
* the protein tree
* @throws ClassNotFoundException exception thrown whenever an exception occurred while
* deserializing an object
* @throws MzMLUnmarshallerException exception thrown whenever an exception occurred while reading
* an mzML file
*/
public static PeptideAssumption getBestHit(
String spectrumKey,
ArrayList<PeptideAssumption> firstHits,
HashMap<String, Integer> proteinCount,
SequenceMatchingPreferences sequenceMatchingPreferences,
ShotgunProtocol shotgunProtocol,
IdentificationParameters identificationParameters,
PeptideSpectrumAnnotator spectrumAnnotator)
throws IOException, InterruptedException, SQLException, ClassNotFoundException,
MzMLUnmarshallerException {
if (firstHits.size() == 1) {
return firstHits.get(0);
}
Integer maxProteins = 0;
ArrayList<PeptideAssumption> bestPeptideAssumptions =
new ArrayList<PeptideAssumption>(firstHits.size());
for (PeptideAssumption peptideAssumption : firstHits) {
for (String accession :
peptideAssumption.getPeptide().getParentProteins(sequenceMatchingPreferences)) {
Integer count = proteinCount.get(accession);
if (count != null) {
if (count > maxProteins) {
maxProteins = count;
bestPeptideAssumptions.clear();
bestPeptideAssumptions.add(peptideAssumption);
} else if (count.equals(maxProteins)) {
bestPeptideAssumptions.add(peptideAssumption);
}
}
}
}
if (bestPeptideAssumptions.size() == 1) {
return bestPeptideAssumptions.get(0);
} else if (!bestPeptideAssumptions.isEmpty()) {
firstHits = bestPeptideAssumptions;
bestPeptideAssumptions = new ArrayList<PeptideAssumption>(firstHits.size());
}
MSnSpectrum spectrum = (MSnSpectrum) SpectrumFactory.getInstance().getSpectrum(spectrumKey);
int maxCoveredAminoAcids = 0;
AnnotationSettings annotationPreferences = identificationParameters.getAnnotationPreferences();
for (PeptideAssumption peptideAssumption : firstHits) {
Peptide peptide = peptideAssumption.getPeptide();
SpecificAnnotationSettings specificAnnotationPreferences =
annotationPreferences.getSpecificAnnotationPreferences(
spectrum.getSpectrumKey(),
peptideAssumption,
identificationParameters.getSequenceMatchingPreferences(),
identificationParameters.getPtmScoringPreferences().getSequenceMatchingPreferences());
HashMap<Integer, ArrayList<IonMatch>> coveredAminoAcids =
spectrumAnnotator.getCoveredAminoAcids(
annotationPreferences,
specificAnnotationPreferences,
(MSnSpectrum) spectrum,
peptide);
int nAas = coveredAminoAcids.size();
if (nAas > maxCoveredAminoAcids) {
maxCoveredAminoAcids = nAas;
bestPeptideAssumptions.clear();
bestPeptideAssumptions.add(peptideAssumption);
} else if (nAas == maxCoveredAminoAcids) {
bestPeptideAssumptions.add(peptideAssumption);
}
}
if (bestPeptideAssumptions.size() == 1) {
return bestPeptideAssumptions.get(0);
} else if (!bestPeptideAssumptions.isEmpty()) {
firstHits = bestPeptideAssumptions;
bestPeptideAssumptions = new ArrayList<PeptideAssumption>(firstHits.size());
}
double minMassError = identificationParameters.getPeptideAssumptionFilter().getMaxMzDeviation();
if (minMassError == -1.0) {
minMassError = identificationParameters.getSearchParameters().getPrecursorAccuracy();
}
for (PeptideAssumption peptideAssumption : firstHits) {
double massError =
Math.abs(
peptideAssumption.getDeltaMass(
spectrum.getPrecursor().getMz(), shotgunProtocol.isMs1ResolutionPpm()));
if (massError < minMassError) {
minMassError = massError;
bestPeptideAssumptions.clear();
bestPeptideAssumptions.add(peptideAssumption);
} else if (massError == minMassError) {
bestPeptideAssumptions.add(peptideAssumption);
}
}
if (bestPeptideAssumptions.size() == 1) {
return bestPeptideAssumptions.get(0);
} else if (bestPeptideAssumptions.isEmpty()) {
bestPeptideAssumptions = firstHits;
}
HashMap<String, PeptideAssumption> sequenceToPeptideAssumptionsMap =
new HashMap<String, PeptideAssumption>(bestPeptideAssumptions.size());
for (PeptideAssumption peptideAssumption : bestPeptideAssumptions) {
sequenceToPeptideAssumptionsMap.put(
peptideAssumption.getPeptide().getSequence(), peptideAssumption);
}
ArrayList<String> sequences = new ArrayList<String>(sequenceToPeptideAssumptionsMap.keySet());
Collections.sort(sequences);
return sequenceToPeptideAssumptionsMap.get(sequences.get(0));
}
/**
* Indicates whether the match designated by the match key validates the given item using the
* given comparator and value threshold.
*
* @param itemName the name of the item to filter on
* @param filterItemComparator the comparator to use
* @param value the value to use as a threshold
* @param spectrumKey the key of the match of interest
* @param peptideAssumption the assumption to validate
* @param identification the identification objects where to get identification matches from
* @param identificationFeaturesGenerator the identification feature generator where to get
* identification features
* @param identificationParameters the identification parameters used
* @param peptideSpectrumAnnotator the annotator to use to annotate spectra when filtering on PSM
* or assumptions
* @return a boolean indicating whether the match designated by the protein key validates the
* given item using the given comparator and value threshold.
* @throws java.io.IOException exception thrown whenever an exception occurred while reading or
* writing a file
* @throws java.lang.InterruptedException exception thrown whenever a threading issue occurred
* while validating that the match passes the filter
* @throws java.lang.ClassNotFoundException exception thrown whenever an error occurred while
* deserilalizing a match
* @throws java.sql.SQLException exception thrown whenever an error occurred while interacting
* with a database
* @throws uk.ac.ebi.jmzml.xml.io.MzMLUnmarshallerException exception thrown whenever an error
* occurred while reading an mzML file
* @throws org.apache.commons.math.MathException exception thrown whenever an error occurred while
* doing statistics on a distribution
*/
public boolean isValidated(
String itemName,
FilterItemComparator filterItemComparator,
Object value,
String spectrumKey,
PeptideAssumption peptideAssumption,
Identification identification,
IdentificationFeaturesGenerator identificationFeaturesGenerator,
IdentificationParameters identificationParameters,
PeptideSpectrumAnnotator peptideSpectrumAnnotator)
throws IOException, InterruptedException, ClassNotFoundException, SQLException,
MzMLUnmarshallerException, MathException {
AssumptionFilterItem filterItem = AssumptionFilterItem.getItem(itemName);
if (filterItem == null) {
throw new IllegalArgumentException(
"Filter item " + itemName + "not recognized as spectrum assumption filter item.");
}
String input = value.toString();
switch (filterItem) {
case precrusorMz:
Precursor precursor = SpectrumFactory.getInstance().getPrecursor(spectrumKey);
Double mz = precursor.getMz();
return filterItemComparator.passes(input, mz.toString());
case precrusorRT:
precursor = SpectrumFactory.getInstance().getPrecursor(spectrumKey);
Double rt = precursor.getRt();
return filterItemComparator.passes(input, rt.toString());
case precrusorCharge:
Integer charge = peptideAssumption.getIdentificationCharge().value;
return filterItemComparator.passes(input, charge.toString());
case precrusorMzErrorDa:
precursor = SpectrumFactory.getInstance().getPrecursor(spectrumKey);
SearchParameters searchParameters = identificationParameters.getSearchParameters();
Double mzError =
Math.abs(
peptideAssumption.getDeltaMass(
precursor.getMz(),
false,
searchParameters.getMinIsotopicCorrection(),
searchParameters.getMaxIsotopicCorrection()));
return filterItemComparator.passes(input, mzError.toString());
case precrusorMzErrorPpm:
searchParameters = identificationParameters.getSearchParameters();
precursor = SpectrumFactory.getInstance().getPrecursor(spectrumKey);
mzError =
Math.abs(
peptideAssumption.getDeltaMass(
precursor.getMz(),
true,
searchParameters.getMinIsotopicCorrection(),
searchParameters.getMaxIsotopicCorrection()));
return filterItemComparator.passes(input, mzError.toString());
case precrusorMzErrorStat:
searchParameters = identificationParameters.getSearchParameters();
precursor = SpectrumFactory.getInstance().getPrecursor(spectrumKey);
mzError =
peptideAssumption.getDeltaMass(
precursor.getMz(),
identificationParameters.getSearchParameters().isPrecursorAccuracyTypePpm(),
searchParameters.getMinIsotopicCorrection(),
searchParameters.getMaxIsotopicCorrection());
NonSymmetricalNormalDistribution precDeviationDistribution =
identificationFeaturesGenerator.getMassErrorDistribution(
Spectrum.getSpectrumFile(spectrumKey));
Double p;
if (mzError > precDeviationDistribution.getMean()) {
p = precDeviationDistribution.getDescendingCumulativeProbabilityAt(mzError);
} else {
p = precDeviationDistribution.getCumulativeProbabilityAt(mzError);
}
return filterItemComparator.passes(input, p.toString());
case sequenceCoverage:
SpectrumFactory spectrumFactory = SpectrumFactory.getInstance();
MSnSpectrum spectrum = (MSnSpectrum) spectrumFactory.getSpectrum(spectrumKey);
Peptide peptide = peptideAssumption.getPeptide();
AnnotationSettings annotationPreferences =
identificationParameters.getAnnotationPreferences();
SpecificAnnotationSettings specificAnnotationPreferences =
annotationPreferences.getSpecificAnnotationPreferences(
spectrum.getSpectrumKey(),
peptideAssumption,
identificationParameters.getSequenceMatchingPreferences(),
identificationParameters
.getPtmScoringPreferences()
.getSequenceMatchingPreferences());
HashMap<Integer, ArrayList<IonMatch>> matches =
peptideSpectrumAnnotator.getCoveredAminoAcids(
annotationPreferences,
specificAnnotationPreferences,
(MSnSpectrum) spectrum,
peptide);
double nCovered = 0;
int nAA = peptide.getSequence().length();
for (int i = 0; i <= nAA; i++) {
ArrayList<IonMatch> matchesAtAa = matches.get(i);
if (matchesAtAa != null && !matchesAtAa.isEmpty()) {
nCovered++;
}
}
Double coverage = 100.0 * nCovered / nAA;
return filterItemComparator.passes(input, coverage.toString());
case algorithmScore:
Double score = peptideAssumption.getRawScore();
if (score == null) {
score = peptideAssumption.getScore();
}
return filterItemComparator.passes(input, score.toString());
case fileNames:
return filterItemComparator.passes(input, Spectrum.getSpectrumFile(spectrumKey));
case confidence:
PSParameter psParameter = new PSParameter();
psParameter =
(PSParameter) identification.getPeptideMatchParameter(spectrumKey, psParameter);
Double confidence = psParameter.getProteinConfidence();
return filterItemComparator.passes(input, confidence.toString());
case validationStatus:
psParameter = new PSParameter();
psParameter =
(PSParameter) identification.getPeptideMatchParameter(spectrumKey, psParameter);
Integer validation = psParameter.getMatchValidationLevel().getIndex();
return filterItemComparator.passes(input, validation.toString());
case stared:
psParameter = new PSParameter();
psParameter =
(PSParameter) identification.getPeptideMatchParameter(spectrumKey, psParameter);
String starred;
if (psParameter.isStarred()) {
starred = FilterItemComparator.trueFalse[0];
} else {
starred = FilterItemComparator.trueFalse[1];
}
return filterItemComparator.passes(input, starred);
default:
throw new IllegalArgumentException(
"Protein filter not implemented for item " + filterItem.name + ".");
}
}